Our purpose is to extend our comprehensive assessment of the neurobehavioral status of diabetic adolescents. Our research during the past 5 years has demonstrated that adolescents who develop diabetes before the age of 5 have a much higher risk of manifesting significant neuropsychological impairments than those diagnosed after that age. Because the test profile of our early onset group is so similar to that seen in brain-damaged children, we have hypothesized that their deficits are organigenic, and are secondary to having experienced serious episodes of hypoglycemia early in life. On the other hand, our later onset subjects show only mild motor slowing on tests requiring speed and accuracy, and we suspect that impairment reflects either a more cautious response style, or a transient reduction in mental efficiency resulting from unusually high or low blood glucose levels at the time of testing. These observations have led us to propose 3 studies which analyze in more detail the qualitative and quantitative differences between early and later onset diabetic adolescents. (1) To examine changes in neuropsychological functioning over time, we plan to conduct a longitudinal study of diabetic adolescents over a 5 year period. (2) To test the hypothesis that early and later onset diabetic subject have very different ways of processing information, we plan to examine the types of strategies used by our subjects when solving complex verbal and nonverbal problems. (3) To test the hypothesis that many early onset diabetic subjects may have suffered significant brain dysfunction, we shall record EEGs from our subjects, and carry out both clinical interpretations and computerized spectral analyses of these recordings.